The present invention relates to a novel method of administration of antigen presenting cells during a multi-administration immune generation protocol. In one embodiment it relates to a novel method of obtaining large quantities of a preferred class of antigen presenting cells, activated B cells.
The traditional approach to generating an immune response was based upon an antigen-antibody reaction, for example by administration of an inactive whole virus. However, our understanding of generating immune responses has increased significantly in recent years. Thus, attention has expanded beyond simply presenting such an antigen. It was recognized that T cells interact most effectively with cells having major histocompatibility complexes (MHC) associated antigens and not soluble antigens. There are two different types of MHC-associated antigens, namely class I and class II. The antigens associated with a particular class of MHC molecule determines the kinds of T cells stimulated by the molecule. Typically, peptide fragments derived from extracellular proteins bind to class II molecules, whereas endogenously synthesized peptides associate with class I molecules.
There are many areas where the traditional presentation of an antigen has not, thus far, generally proven clinically successful such as using tumor-associated antigens, HIV, etc. Thus, considerable attention has focused on understanding the method of generating and regulating immune reactions such as those generated by the MHC-antigens, in order to more appropriately regulate the process.
It was discovered that there is a group of cells that typically process proteins via endocytosis thereby subjecting them to enzymatic and chemical degradation to result in a xe2x80x9cprocessedxe2x80x9d peptide. This peptide will then bind to an MHC molecule which transports it to the surface where it is presented a to T cells for appropriate interaction. Such cells are called antigen presenting cells (APCs). Thus, paradigms have been proposed for the appropriate structure of MHC I and MHC II peptides that are presented. [WO 94/20127; Bartholomew, J. S., et al., Eur. J. Immunol. 24:3175-3179 (1994); Falk, K., et al., Nature 351: 290-296 (1991)]. For example, MHC class I molecules bind preferentially to peptides 8-10 residues, whereas class II molecules bind preferentially to peptides 12-25 residues long. In these peptides, there are certain amino acid residues that are more critical and tolerate only certain amino acids. Similarly, there are limitations on the cells that can serve as APCs for one class of molecules as opposed to another. The number of cells that are suitable class II APCs is substantially smaller than for class I. It would be desirable to have an APC that will be useful with both MHC classes. Although it has been known that a number of different cells naturally can be used as MHC class II APCs, such as mononuclear phagocytes, dendritic cells (DCs), Langerhans cells of the skin, activated B lymphocytes and endothelial cells, considerable attention has focused on using dendritic cells as the APC. The reason for this attention includes its high efficiency in antigen presentation, relative ease of isolation [Mackensen 1995 #56] and relative ease of culturing. For example, DCs are about 5-10 fold more efficient at presenting alloAg than activated B cells. DC can be obtained as stem cell derived DCs, either from bone marrow or peripheral blood stem cells, or peripheral blood derived DCs. Dendritic cells prepared from bone marrow cells [Caux, 1992 #55]; Mackensen, 1995 #56; Szabolcs, 1995 #73; Bernhard, 1995 #74], demonstrate APC function [Nussenzweig, 1980 #12; Tew, 1982 #58; Steinman, 1991 #64; Steinman, 1991 #65], and the knowledge of definition of the culture conditions needed to expand larger numbers of dendritic cells [Mackensen, 1995 #56]; Inaba, 1992 #72] has made this cell population the present choice for use in vaccination strategies [Grabbe, 1995 #48]; Caux, 1995 #49; Young, 1996 #54]. Moreover, it was demonstrated in murine model systems that DCs pulsed with tumor peptide antigens in vitro can induce a T cell mediated tumor specific immune response in vivo [Paglia, 1996 #25]; Cohen, 1994 #51; Zitvogel, 1996 #50; Celluzzi, 1996 #52; Flamand, 1994 #53].
However, these cells have limitations, including diminished long term capacity. Stem cell derived DCs have to be expanded using several different cytokine cocktails. This procedure takes a relatively long time and is cost intensive. A culture period of 35 days under optimized conditions using 7 different cytokines was necessary to obtain 1.7xc3x97107 DCs from a starting population of 1xc3x97106 mononuclear cells of a peripheral stem cell (PBSC) preparation [Mackensen, 1995 #56]. In another study, to generate DCs in vitro from peripheral blood high amounts of GM-CSF and IL-4 were necessary [Romani, 1994 #14]. The yield of DC in that study was about 3-8xc3x97107 DCs from 40-100 ml peripheral blood after 5-8 days of culture, but growth ceased at that time and no further expansion was possible. Another limitation of generating DCs from bone marrow (BM) or peripheral blood (PB) is the decreased ability of long term cultured cells to function as APC [Mackensen, 1995 #56] due to down-regulation of important molecules such as CD80 (B7-1). Yet, another limitation is that DCs cannot be stored long term since they cannot be cryogenically frozen.
It would be desirable to be able to use other cells as APCs, if they could be prepared more efficiently and effectively then dendritic cells.
Typical modes of generating immune reaction involve multiple injections of APCs over a course of administration that takes place over extended periods of time. General protocols require an initial administration and subsequent boosts at intervals ranging from one to two weeks to up to several months. One preferred protocol is to administer the APCs approximately at 1 to 2 weeks intervals, 5 to 10 times, more preferably 6-7 times. However, such protocols effectively limit the use of one draw of dendritic cells to at most 2 administrations before additional blood must be drawn, purified and cultured. Thus, in addition to receiving the boost, a patient will also have about 40-100 cc of blood drawn at each administration, in order to culture and prepare more DC.
It is now well established, that T cells are activated upon recognition of peptide antigen presented by the major histocompatibility complex (MHC) on professional antigen presenting cells (APCs) (signal 1) [Schwartz, 1989 #31; Schwartz, 1990 #32] in combination with costimulation which is mainly provided by members of the B7 family (signal 2), on the APC to the CD28 molecule on the T cells [June, 1990 #29; June, 1994 #30]; Schwartz, 1992 #33; Linsley, 1991 #34; Freeman, 1989 #35; Freeman, 1993 #36]. White dendritic cells, B cells and macrophages are known to function as APCs. However, it is still not clear whether all APCs function in concert during (1) onset, (2) amplification and (3) expansion of an immune response or whether there is a hierarchy of interactions between professional APCs and T cells. In addition, different routes of entry of antigens into the organism and their different origin might influence which APC might play the major role in an immune response. Consequently, being able to use different APCs would be desirable. However, heretofore, when multiple administrations were necessary, the use of DCs was preferable to the use of activated B cells because one would have to administer a greater number of activated B cells then DCs to obtain the same effect. Although B cells are known to proliferate for long periods of time, they are typically outgrown by T cells. Even if one highly purified B cells, one could not get rid of all T cells present. Consequently, in a relatively short period of time, for example, 14 days, a considerable T cell population would be present in the B cell culture rapidly becoming the predominant population. Accordingly, it has not been feasible to use a single culture of B cells for a series of immunizations where multiple administrations were necessary.
We have now discovered a means that permits both the rapid proliferation of activated B cells, their purification from T cells, and their sustained culture thereby permitting the use of a single culture of B cells as the APCs in a multiple administration method of generating an immune response.
We have now discovered that activated B cells can be used to efficiently induce a T cell mediated immune response, thereby modulating an immune response in a multi-administration protocol. Preferably, the B cells help generate an immune response. As a result of our invention the activated B cells fulfill the following criteria: (1) they serve as APCs expressing all necessary molecules for antigen processing and presentation, (2) they are easily accessible, (3) they are expandable in vitro, and (4) they are able to present efficiently a desired peptide, such as a tumor antigen, to T cells. B cells express MHC class I and II and can therefore be used with a wide range of antigens.
While B cells are easily accessible from peripheral blood they have not previously met the criteria for use in the method proposed. Under physiological conditions resting B cells do not express all necessary molecules, such as the costimulatory molecules, to induce a sufficient T cell response [Jenkins, 1990 #37]. Instead, they were shown to anergize T cells [Matzinger, 1994 #38]; Fuchs, 1992 #39; Lassila, 1988 #41]. Only activated B cells express sufficient levels of costimulatory and adhesion molecules to induce T cell proliferation optimally [Schultze, 1995 #28; Boussiotis, 1993 #42]. The major B cell activation pathway is crosslinking of CD40 by CD40L (sometimes also referred to as gp39) that is expressed on activated T cells [Banchereau, 1994 #43]; Armitage, 1992 #44; Clark, 1994 #6064; Hollenbaugh, 1992 #45; Hollenbaugh, 1994 #46]. Normal and malignant B cells can be activated in vitro with transfectants expressing the CD40L. Once activated these CD40-activated B cells (CD40-B) express high levels of MHC molecules as well as adhesion and costimulatory molecules [Schultze, 1995 #281; Ranheim, 1993 #47]. The activated cells are highly efficient antigen presenting cells [Schultze, 1995 #28], and the B cells can be expanded in this culture system [Banchereau, 1994 #43]. However, the methods used to activate B cells also activated T cells. B cells constitute a very small percentage of blood and are difficult, if not impossible, to purify so that no T cells are present. Since, the co-cultured T cells outgrow the B cells, the T cells rapidly become the predominant species being cultured.
We have overcome the problem and discovered a method where normal human B cells can be obtained easily from peripheral blood, do not need to be enriched prior to culture, can be expanded in vitro with a CD40L, e.g., on CD40L transfectants, express all necessary molecules for antigen presentation, and induce proliferation of allogeneic CD4+ and CD8+ T cells. This method involves using an immune suppressor under conditions where it preferentially affects T cells, rather than B cells. Thus, the use of compounds such as cyclosporin A (sometimes also called cyclosporine), mycophenolate mofetil, azathioprine and tacrolimus. More preferably, cyclosporin A. As a result of this method, one can obtain an activated B cell culture that is highly purified. Preferably, at least 85% pure, still more preferably at least 90% pure, even more preferably at least 98% pure and most preferably at least 99% pure.
The cultures can be grown over long periods of timexe2x80x94with increasing activated B cell number. Preferably, at least 2 months, more preferably at least 3 months, still more preferably at least 6 months.
Additionally, the activated B cells can be cryopreserved. Cryopreserved material can be used even 5 years later.
The activated B cells can thus be used as APCs from a single culture over an extended period of time. One can combine the APC with a desired antigen or a cocktail of antigens by a variety of techniques well known in the art. For example, one can present the whole antigen, which is then processed, a specific peptide, nucleic acid encoding the antigen or the peptide. One preferred method is to target a nucleic acid encoding a desired antigen to a particular cellular compartment. Another embodiment involves culturing the APC with the protein or peptide (pulsing) for a sufficient time for the APC to process and present the MHC-bound peptide. Thereafter one administers the active APC to the subject. The magnitude of T cell proliferation is dependent on cell number of APCs administered. Because of the large cell number that can be obtained over time the T-cell reaction can be greater than that induced by DCs. Any appopriate desired target antigen can be used. For example, we have shown that our CD40-B cells can be pulsed with tumor peptide to induce autologous CD8+ T cells to proliferate and kill target cells expressing the antigen. Additionally, the T cell proliferation induced by DCs in the presence of immunosuppressive cytokines is significantly diminished. In contrast, when our CD40-B cells were used under such conditions there was not a reduction in activity.